PROJECT SUMMARY Late onset Alzheimer?s disease (LOAD) is the leading cause of dementia among the middle aged and elderly. It is characterized by progressive loss of memory culminating in complete loss of cognitive function. Pathological manifestations in brain include neuronal loss, gliosis, extracellular amyloid deposits and neurofibrillary tangles. Among genetic risk factors, APOE-?4 is the strongest, but genome association studies (GWAS) have identified and confirmed several additional loci. However, a substantial portion of the genetic source of heritability of LOAD is still unknown. Large-scale sequencing efforts in Alzheimer?s Disease are underway to identify detect putatively causal rare variant (RV) associations that might explain the missing heritability. To detect modest RV effects, large sample sizes are required. Here we propose a powerful approach to study RVs in extended families with large numbers of affected individuals where they are likely to aggregate and have stronger effect sizes. The major goal of this proposal is to harmonize phenotype and whole genome and exome sequencing (WGS and WES) data from ~1000 LOAD families and apply existing and novel family-based analytics to identify LOAD susceptibility variants and loci that can be tested as therapeutic targets. Factors such as a three to five-fold higher incidence rates of LOAD compared to the general population, clustering of putatively deleterious variants, inbreeding, low level of sequencing artifacts and the ability to control for population substructure/admixture using a family analysis design (compared to unrelated case-controls) make these families ideal for identifying LOAD associated genetic risk and protective factors. The efforts in this proposal will be in parallel with and complementary to analyses in the unrelated case-control analyses being conducted on Alzheimer?s Disease Sequencing Project (ADSP) discovery, extension replication and follow-up datasets.